【摘要】 为应对全球变暖和能源危机，国内外以光伏为代表的新能源产业得到了快速发展。然而，光伏发电的间歇性和不确定性给并网和电网系统带来了巨大挑战。精准的光伏功率预测能够为电网系统调度提供依据，进而提高运行稳定性。围绕超短期光伏功率预测技术，从预测精度影响因素、评价指标、预测方法、方法对比分析和总结展望五方面展开讨论。预测方法划分为基于站点监测数据、卫星云图、地基云图和多源数据四类，对每一类数据适用的预测范围、数据特性、常用数据源以及国内外相关技术进行总结分析。方法分析总结了不同数据源的优缺点，并比较了常用预测模型。总结展望指出未来超短期光伏预测亟须解决的关键技术问题。更多还原

【Abstract】 To address the global warming and the energy crisis,the new energy industry,represented by photovoltaics,has experienced rapid development domestically and internationally.However,the intermittency and uncertainty of photovoltaic power generation pose serious challenges to grid and power systems.Accurate photovoltaic power forecasting can provide a solid basis for the grid system dispatch,thus improving the operational stability.This review focuses on ultra-short-term photovoltaic power forecasting technology and discusses five aspects,including factors influencing forecasting accuracy,evaluation metrics,forecasting methods,comparative analysis of methods,and a summary and outlook. The forecasting methods are further categorized into four types based on monitoring data at the site,the satellite images,the sky images,and multi-source data.For each data category,a summary analysis is provided including applicable forecast scope,data characteristics,commonly used data sources,and relevant technologies domestically and internationally. The method analysis compare commonly used prediction models and summarized the advantages and disadvantages of different data sources.The concluding outlook points out the key technical challenges that need to be addressed in future ultra-short-term photovoltaic forecasting.更多还原